Descriptions of the Audio Examples

This appendix provides short desciptions for the accompanying audio examples. The text provided here is similar to the spoken words on tape preceeding each example.

Examples for Chapter 4

Ex. 16

A Shepard tone with 8 partials, starting at 32 Hz, is played first at normal speed and then twice as fast. The partials of this sound are geometricaly related to each other by a factor of 2. The claim is that the percieved pitch of the sound remains the same even though it is being played twice as fast. The example is played twice.

Ex. 17

A Shepard tone similar to the previous example, except with partials which are gemetrically related to each other by a factor of 2.12, is played first at normal speed and then twice as fast. The claim is that, paradoxically, the percieved pitch of the sound is decreased by a half step when the sound is played twice as fast. The example is played twice.

Ex. 18

This example is the resynthesis of the first 30 seconds of the extracted ``top voice'' from J. S. Bach's 3rd Brandenburg concerto.

Examples for Chapter 5

Ex. 19

This example illustrates the result of self-similar synthesis for a binary segmentation, and frequency factors of 1 and 2 is played. A graph of the frequency fluctuation of this example can be seen in figure 5-1.

Ex. 20

This example illustrates the effect of a trinary segmentation with frequency factors of 1 and 1.5. A graph of the frequency fluctuation of this example can be seen in figure 5-2.

Ex. 21

This example illustrates the effect of all-level synthesis for a trinary segmentation. The basic structure of this example is the same as the previous example except that all the levels are synthesized and added together, and lower amplitude factors are used.

Ex. 22

This example illustrates the effect of all-level synthesis with binary segmentation and frequency factors of 1 and 2.

Ex. 23

This example illustrates the effect of using a sinusoid window with time segmentation of 1 to 20. The self-similarity of this signal is illustrated in figure 5-3.

Ex. 24

This example is a 5 second version of the previous example and it illustrates the scalability of the synthesis process.

Ex. 25

This example illustrates the effect of an unequal trinary segmentaion. The different partials of this example are harmonically related.

Ex. 26

This example illustrates the effect of magnifying the structures of a short sound by layering. The amplitude ``window'' for this example is a segment of a spoken word which is played before the example.

Ex. 27

This example is similar to the previous example except that its amplitude ``window'' is a long cello melody. The original cello sound is played before the example.

Ex. 28

This example illustrates the effect of layering many transposed copies of a looped piano note.

Ex. 29

This example illustrates how the system can be used for creating rhythmical pieces. The amplitude ``window'' is a segment of a sampled powertom drum sound which is played before the example.

Ex. 30

This example illustrates how a compound rhythm could be created and how multiple instruments could be used in a piece. This example uses three different segments of sampled sounds as instruments which are played before the example.